iii

68 Tr a nsportation Research Record 1032

The Effect of Vegetation Transpiration on the

'' o1r1 .. ..._,_...__

Soil Foundation

CHEN LIN and TIAN KAIMING

ABSTRACT

Vegetation transpiration may significantly decrease the moisture content of active soils causing soil shrinkage and deformation. Severe damage to engineer­ing structures may result. This phenomenon has been identified in reports from South Africa, Australia, and China as a cause of structural damage. This paper contains analyses of data collected in Mengzi County and other regions in China. The relationship between veget~tive transpiration rates and deformation of high void ratio expansive soils is examined, and a formula for calculating estimated maximum contraction is presented. Also, current practices for dealing with active soils are discussed.

DAMAGE TO ENGINEERING

Plants that cover the earth surface are termed vege­tation. Water and nutrients are assimilated by the plant root system and delivered to the leaf surface, stems, and other plant parts via capillary tubes. Only 1 percent of the absorbed water is consumed for the growth of the plants: the balance is dispersed into the atmosphere in the form of vapor by stomata on the leaf surface.

This process of water evaporation through the plant organ is called transpiration. It is shown by the measured data that different types of trees have different transpiration rates (Table 1). Among the more common trees, the jar rahs have the greatest transpiration.

ll«i>d '"'' onP. of the measures to lower the ground­water table in Australia, the jarrahs are often com­pared to pumps. In some reg ions, because of the large loss of water from the foundation soil, which was caused by the transpiration of jarrahs, irregular

shrinkage deformation of the foundation occurred. This caused serious engineering failures to take place.

Investigation indicated that in the expansive soil regions of Yunnan and Guangxi Province, China, building damage phenomena due to the vegetation effect are rather common (Table 2) •

In these regions, building damage rates in places with growing trees are higher than in places without trees, which obviously shows the close relationship between building damage and vegetation. In Mengzi, for example, most buildings on expansive soil foun­dations not surrounded by trees remained in good condition. Only slight damage caused by uneven ex­panding and shrinking deformation of the foundation soil occurred. The deformation amplitude is not more than 15 mm and the ~·.' idth of wall crat:!k is less t ha n 10 mm. However, 90 percent of the buildings close to jarrahs and grevillea were seriously damaged. The maximum de f ormation amplitude of foundations buildings near trees is up to 187.2 mm, and

of the

TABLE l Transpiration of Common Species of Trees

African Glossy Chinese Barren Jarrah Eucalyptus Grevillea Privet Parasol Land

Transpiration kg/day South Africa 1,090 454 342 61.2 0.54 Mengzi 432 68

TABLE 2 Investigative Data of Houses Destroyed by Vegetation Influence

Tree-Growing Site Non-Tree Site

Rate Rate Rate No. Houses No . Houses Destroyed No. Houses No. Houses Destroyed No. Houses No. Houses Destroyed

Region Investigated Destroyed (%) Completed Destroyed (%) Completed Destroyed (%)

Mengzi 145 109 75.l 9 100 91.7 27 9 25 .0 Jianshui 56 44 78.5 4 17 80.9 8 27 77.1 Ningming 53 35 66.0 3 30 90.9 15 5 25.0 Nanning 46 25 54.3 13 2 1 61. 7 8 4 33.3 Gui County 79 39 49.3 30 28 48.2 9 12 57.1 Yanshan 42 25 59.5 3 12 80.4 14 13 48.1 Kunming 34 20 58.8 2 12 85.7 12 8 40.0

... iiiii

Chen and Tian

maximum width of wall crack is 240 mm. Partial wall inclination reaches 18.9 percent (Table 3). In this area the greater the afforestation the more serious the damage. In the vicinity of the trees, structural walls nearest the trees exhibit cracks, deformation, or foundation damage.

A 36-m long, single-story building with a jarrah near one end of the front wall and no other trees nearby experienced par ti al shrinkage of the found a­t ion soil near the tree. The brick work was twisted, the side line and foundation beam of the back wall was bent in the form of an arc with the tree as a circle center and the brick side line of the front wall squeezed in undulation form (Figure 1). Figures 2-6 show the damage to houses under the influence of trees. These figures also show the corridor brick pillar of some houses displaced 10 cm horizontally toward the trees. Figure 7 shows the depth of the tree roots near the houses. In some houses, rein­forced concrete foundation beams with 40 x 40 cm"

Brick Sideline in Undulation Form

-~ Plan of Twisted Brick Work

FIGURE I Orthogonal projection of twisted house near jarrah.

FIGURE 2 Horizontal displacement of corridor brick pillar.

69

FIGURE 3 Crack on gable of building near jarrah.

FIGURE 4 Inclination of external wall.

FIGURE 5 Collapse of roof.

70 Transportation Research Record 1032

TABLE 3 Statistical Data of the Relationship Between Building Deformation and Vegetation

Building No. Found ation Type

Embedded Fo undation Depth (m ) Constru ction Type

1-5 Strip foo ting on sand cushion I.S O 1.20 3.95 3.95 3.95

Brick and wood, one-story Brick and wood, one-story Brick an d concrete, three-stories Brick an d concrete , three-stories Brick and co ncrete 1 three-sto ries

1-7 Pier found ation 11-1 Pier foundation 11-3 Pier found ation 11-5 Pier fo und ation

11-2 Pier found ation 3.95 Brick and co ncrete, three-stories

Ii-20 Strip fo o ting 0.80 Brick and concrete, one-story

11-16 Pedestal pile 3.00 Brick and concrete 1 two-stories

IV-6 Strip footi ng 0.80 Brick and co ncret e, two-sto ries

Jiang-shuidi Hotel Strip foo ting 0.80 Brick and concrete, one-story

I-Auditorium Strip foo ting 0.80 Brick and wood

III-24 Strip footi ng 0.80 Brick and wood, two-stories

30ue to the high solid rigid ity o f building no crack occurred .

FIGURE 6 Breakage of foundation and foundation beam caused by crack through soil under foundation.

cross sections were broken by stretching. Some roofs caved in, exposing the rooms to the sky, and some external walls collapsed. Many houses collapsed without demolition. The remaining houses were no longer usable. They had suffered damage as s evere as that caused by an e a rthquake.

FOUNDATION DEFORMATION

Vegetation transpiration causes water loss of foun­dation soils and forms a limited shrunken section-­the main reason that adjacent buildings suffer dif­ferential settlement of foundation elements. The data observed in the Mengzi region indicated that the lower limit of water content variation of foun­dation soil is low in an area influenced by jarrahs, and the water c ontent variation is wide. The ampli­tude between expans ive and deformation is substan­tial. At a shringage depth o f 3 m, amplitude of water content variation of f ounda t i on soil in forest zones is 2 to 3 times that in barren land, and the

FIGURE 7 Extended depth of tree roots, which is equal to the height of the t ree and observed from the excavated cross section.

deformation amplitude is 8 to 20 times as great (Table 4). It is evident that the effect of vegeta­tion transpiration on foundation soil is greater than the effect of weather evaporation.

In vegetated areas, foundation deformation is related to seasonal weather variations. Measuring mark deformation curves show regular variation as they move upward in the rainy season and downward in the dry season (Figure 8). The measuring mark of those that drop greatly also rise significantly. Thos e that drop slightly rise only slightly. This indicates that the foundation soil experiences reversible expansive and shrinking deformation, which alternate endlessly as humidity and soil moisture increase and decrease. Because the observed contraction of foundation soil was greater than the expanding deformation in a period of 1 1/2 yr, the curve of relationship between measuring mark and time inclines downward continuously, and the maximum sinking value at 1 m depth is up to 1 28 mm. Elimina­tion of the tree roots causes the curve to rise continuously. In 8 months the rising amplitude

... ...

Chen and Tian

Effect

Vegetation Condition

Non-tree Non-tree Non-tree Non-tree Rows of jarrahs

Rows of jarrahs

Rows of jarrahs

Rows of jarrahs

Independent jarrah

Rows of jarrahs

Rows of jarrahs

Rows of jarrahs

Distance to Tree (m)

5-8

8-10

3-10

3-5

3-6

3-5

3-2

3-5

Maximum Deformation Amplitude Near Tree (mm)

35.0

69.2

83.l

46.8

187.2

66.8

Maximum Deformation Amplitude (mm)

7.6

11.4

Building Condition

Good Good Good Good 2.94 percent inclined to direction

of tree' 6.6 percent inclined to direction

of tree• Medium damage, wall crack width

3-5 cm Slight damage, wall crack width 2-S

cm Serious damage, wall crack width

8-20 cm Medium damage, wall crack width

2-3 cm Serious damage, gable and roof

truss collapse Serious damage, gable and roof

truss collapse

TABLE4 Jarrah Deformation Amplitude-Forest Area and Barren Land

Depth(m)

Barren Forest Barren Forest Barren Land, Area, Land, Area, Land, 1.0 1.0 1.5 1.5 2.0

Water content Rainy season 36 38 36 38 37 Dry season 32 26 30 26 31 Amplitude of variation 4 12 6 12 6

Deformation amplitude of measuring mark, mm 12.6 105.1 11.9 99.1 7.7

measuring mark at 1 m depth 20.0

10.0 ··············-~--.. ~-·~·~- ~ ...

Forest Barren Area, Land, 2.0 2.5

36 41 26 35 10 6

96.2 3.8

0.0 ......

-i---.:_==-=-~=-=.=-=.-:::.:_!::_:_=:::;;_:--..l------'------'---1 ~istance (m)

-·········' -10.0

-20.0

'E ·30.0

! • 'B

20.0

. t! Q. 10.0 E < c 0.0 .2

i -10.0 15

~ -20.0

-30.0

·40.0

10.0

0.0

·10.0

············• · .. '-._,_ .. -..... -.... , ... ___ ,,,,. .. •••••• ••• ••• •

5.0 10.0 15.0 20.0 25.0 measuring marl< at 3 m de th

......................................................... ········ ............. ..

:; .. ;;;; ................ ;--~ .. -.. ,

5.0

.. '­.. -, . -10.0 15.0 20.0

measuring mark at 5 m depth 25 .0

L Distance (m)

-t--~·~;.~;·;.~~~ .. ~~~-~~~~-~~-~-~~-.~··~·~··~·~--~·~··i·,i·~~-~-·~-~--~-i~'i:~~~~~:~~~--~~~~~~~-~~~j·;~~·~:~F~~~~~·;··;~;··ii;.:----41 ~;..,_ (ml

5.0 10.0 15.0 20.0 25.0

Forest Barren Area, Land, 2.5 3.0

41 39 30 35 11 4

74.5 3.2

E:3 September 1981

1 ....... <J July 1981

~ Mey1981

E3 March 1981

1!i!!9 January 1981

FIGURE 8 Deformation curves of measuring marks at different depths under the influence of jarrahs.

71

Forest Area, 3.0

37 29

8

42.8

72 Transportation Research Record 1032

79 80 81 82 Time (month) I

11 , 3 5 7 9 3 5 7 9 34 7

Or I 11 I ll 1 1~1 11 •• <..: I I I J I

••- -.__ before 5 m

"' Q) lU "•• '-. trenching j .

•. -......_ _ - after trenching "Cl :J

.<::: c. E <(

40 •• .. I ·. ' . .. •••• "\. 1 3 m ••••• ••• ' I 1·-~,, __ ;.,~-.-.·

60

c: 0

80 . .. . •'\. 1··· -~ 100 E

•• .. I • · .. '\.. ,,/' .. •,. Y ,.••1 m ·'8.·· 0

~ 120

FIGURE 9 Curve of relationship between measuring mark and time.

reached 66 mm (Figure 9) • The reversible deformation character of the expansive foundation soil and the formation of partial "dry shrunk area" of the foun­dation soil caused by vegetation transpiration were obviously indicated by the previously mentioned ef­fect of return rising.

Vegetation transpiration unbalanced the water condition in the soil, thus causing the formation of a new humidity field as well as humidity variation within the foundation soil in the dry shrunk area (Figure 10). Therefore the water suction ability of the foundation soil matrix also varied. Because of

2.0

3.0

6.0

Distance ( m)

6.o I 3.o I s.o

40%

45%

50%

5.0

FIGURE 10 Humidity isogram of jarrah influence area.

this difference, the water in the soil transfers continuously to the tree roots. After cutting a lime trench between the house and tree in the jarrah-af­fected area, both the C and D measuring mark groups at the side of the house had 2 yr rising deformation, whereas sinking deformation occurred on the A and B measuring mark groups (Table 5 and Figure 11). On the house side, transfer of the soil water to the ~ree had been stopped after separation of the faun-

dation soil from the influence of the tree roots. Before trenching, the water content of the founda­tion soil inside the house was 31 percent, after trenching it increased to 41 percent. Thus, the C

and D measuring mark groups rose rapidly, and the D group, which was inside the house, had the largest rising amplitude--the maximum rising rate reached 40 mm/month. On the tree side, due to the decrease of dry shrunk area, water consumption from the foun­dation soil for vegetation transpiration increased, and the foundation soil showed substantial de­formation.

In the foundation soil affected by the tree, the sphere to be bounded within the deformation-in­fluenced radius horizontally and depth vertically is called dry shrunk area. Its scope depends on the species, age, trunk diameter, and root growing con­dition of the trees. In the Mengzi region, a growing jarrah 10 yr old with a 20- to 40-cm diameter trunk caused the ground surface to be deformed with 15 to 37 mm amplitude, and it has an influence sphere of about B m radius (Figure 12). A deformation curve of 72 measuring marks in different depths near a fully grown jarrah 17 m high with a trunk diameter of about 50 cm shows that deformation differential of measur­ing marks in different parts of the dry shrunk area is great (Figure 13) • It was discovered after exca­vation that the humidity of the part of the founda­tion soil where the tree hair reaches varies greatly as well as the deformation amplitude (Figure 7). The deformation amplitude of the measuring mark with distance to tree beyond 25 m was obviously reduced and nearly equal to the deformation amplitude of measuring mark in the area free of trees. This indicates that the deformation- influenced radius under the jarrah is 25 m, which approximately equals 1.4 times the tree height. As a rule the deformation amplitude of measuring mark in the dry shrunk area decreases progressively with the increase in depth (Figures Band 13). Assuming the depth of measuring

marks with deformation amplitude < 10 mm to be the tree-influence depth, the influenc;; depth of jarrahs in the dry shrunk area is 3.5 to B m. The farther the spot from the tree, the smaller the influence depth will be (broken line in Figure 14) • The in­fluence depth measured from the mark with 25 m dis­tance from the tree is 3.5 m. This is similar to the weather influence depth in space without trees.

INFLUENCE FACTORS

As mentioned previously, the main factor that af­fects the foundation soil deformation in vegetation areas is vegetation transpiration and the expanding and shrinking character of the soil. Beyond the species of tree, the amount of transpiration is

... -

Chen and Tian

"' Cn 3

!" 0 3

"' 0 3

rr= =-=---=--=--=--=--=--=-~ "=-I

ii I D·. ~ IL c · · · · L~-= = :__ --=~:= .=: ~-~-;:-"l

trench A· •• ,

~ m-i~ \!J§I ® W jarrahs

1 = 500

FIGURE 11 Location of measuring marks at different depths on both sides of lime trench.

73

related to the local weather in addition to the restrictions of the hydrogeology and engineering geology condition of the site. Rothenberg believes that the relative humidity of the environment can restrain vegetation transpiration (1). The higher the relative humidity, the lower -the vegetation transpiration.

Thus, the effects of vegetation transpiration on deformation of foundation soil in humid climates are smaller than in dry climates. The effects of vegeta­tion transpiration on water content variation and deformation of foundation soil are small on the site that has a thick earth cover or a high groundwater table. But on the site that has a thin earth cover and a low groundwater table, because of the trans­piration influence caused by water suction of the tree root, the water loss of the foundation soil is

TABLE 5 Deformation Amplitude of Measuring Marks at Both Sides of Lime Trench

Measuring Mark Group A Deformation (mm), Depth (m) Rising Sinking

1.0 0.0 -21.2 2.0 0.0 -21.1 2.5 0.0 -13.0 3.0 0.0 -8.8

Group B

Amplitude Rising

21.2 0,0 21.1 0.0 13.0 0.0

8.8 0.0

Sinking

- 22.6 - J 7.5 - 10.8 - 5.3

Amplitude

22.6 17.5 10.8 5,3

space • 7

Group C

Rising

9.3 9.0 6.8 4.2

Sinking

o.o 0.0 0.0

-1.1

space • 8

Amplitude

9.3 9.0 6.8 5.3

Group D

Rising

43.2 23.4 10.4

8.4

space • 8

Sinking Amplitude

0.0 43.2 0.0 23.4 0,0 10.4 0,0 8.4

f - forest

-20.0-'---------- --------------------1 4M BM 13M

I I

5.0 0.0~ 10.0 ------------

15.0

21M 2BM 36M I I

FIGURE 12 Observed cross section of surface deformation near jarrahs in Mengzi region.

20.0

c: 40.0 0 -~

E so.o ~ ., Cl 80.0

100.0

Distance (m)

2.5 5.0 7.5 10.0 12.5 15.0 20.0 25.0

140.0....L.-----------------------------------------' ~ J, ~ b ~ Measuring Lri .; ..; N ,..: Mark Depth

FIGURE 13 Relationship between deformation amplitude of measuring mark and distance to tree.

74 Transportation Research Record 1032

Distance to Tree (m) 10.0 12.5 15.0 20.0 25.0

20

E ~ 40 'C. QI

0 ..........

••••••.•••••••• 5 .. 0 m

J •• • 3.5 m ..

······s.sm .. ··············s.5 ~-···s.o m

... ·'· 7.0 m

60

80 8.0m

FIGURE 14 Influence depth of measuring marks at different distances to tree.

high; therefore, substantial shrinking deformation can easily occur.

It should be pointed out that vegetation trans­piration is an external factor that causes the humidity variation of the foundation soil. Only the expanding and shrinking character of the soil is the essential factor of deformation of foundation soil. Under the influence of vegetation transpiration, the foundation soil with high void ratio and strong expanding character will have a low lower limit of water content variation and remarkable shrinking deformation. For example, in the Mengzi region, the humidity coefficient of foundation soil is Km = 0.6, free expansion ratio is ep8 = 81 percent, and the natural void ratio is e 0 = 1.15. Because of the small humidity coefficient of foundation soil, dry weather, high vegetation transpiration, as well as high void ratio and strong expanding character, the effect of vegetation on deformation of foundation soil in the Mengzi region is rather great (Tables 3 and 4).

The deformation amplitude of measuring marks adjacent to the tree is up to 105.1 mm, that of parts of the house close to the tree reaches 187.2 mm. In this same area, however, the nonexpansive foundation soil is ep8 = 30 percent and e0 = 1.41, and the deformation amplitude of measuring marks near the jarrah at a 1-m depth is only 30 mm.

Hefei has a subhumid climate (Km = 0.8, eF8 = 64 percent, and e 0 = 0.65) i deformation amplitude of parts of the house near the tree is 7.82 to 11.3 mm. In the Chengdu area (Km = 0,9, ep8 = 61 percent, and e 0 = 0.64), deformation amplitude of parts of the house close to the tree is 18 mm. This value is ap­proximately equal to the deformation amplitude of an area without treesi therefore it is evident that the deformation of foundation soil is not affected by the tree. It is evident from the preceding examples that the deformation of foundation soil is related to Km in inverse ratio and to e0 in direct ratio. Therefore the effect of vegetation on buildings with high void ratio expansive foundation soil is remark­able in dry and sub-dry regions in China.

The possible shrinkage rate (81 ) of the founda­tion soil under the influence of weather and soil character may be calculated from the formula given below when the following conditions are satisfied:

1. Foundation soil is beyond the influence of groundwater and eF8 ~ 40 percent; and

2. Deformation of foundation soil refers to the shrinkage that is caused by the decrease of water content, and the void ratio variation of soil is in direct proportion to the variation of water content.

where

81 possible shrinking detormation rate at 1-111 depth of foundation soil because in measuring on-site the greatest deformation is obtained at 1-m depth in the vegetation area; there­fore, only the deformation amplitude of foun­dation soil at 1-m depth is calculated, and the average rate of deformation within the 1-m depth of foundation soil is taken for calculating ratei

e0 ~ natural void ratio of soil; em void ratio of soil in which water content

is minimum Wm, ~ Rl 0.028 Wm+ 0.041 may be taken in lack of measuring data, and

Wm minimum water content that depends on character of soil, weather, and humidity of foundation (~_).

Possible shrinkage rates of foundation soil in 40 places in regions with high vegetation transpiration have been calculated according to the preceding for­mula. The results are compared with the actual de­stroyed condition of the corresponding houses. The following conclusions can be drawn: in areas in which 81 is inversely proportional to Km and Uire~tly propurtional Lo e

0, Krn ~ 0.9, e 0 ..:_ 0.7, or

81 < 10 percent, damage is not a result of vegeta­tion; in areas in which Km< 0.9, e 0 > 70.7, or 81 ~ 10 percent, damage to the houses is related to the vegetation, and the greater the 81, the greater the vegetation's influence (Table 6 and Figures 15, 16). Thus the conclusion may be drawn that 81 may be used as a quantitative index in predicting and estimating the effect of vegetation on foundation soil deformation. In the area in which 81 ~ 10 per­cent, the possible engineering damage caused by the influence of trees with high transpiration should be taken into account, whereas in the area in which 81 < 10 percent the influence of trees can be discounted.

TREATMENT

The vegetation influence presenting measurements and afforestation experience in Mengzi and other regions can be summarized as follows:

1. When building houses in areas expected to be influenced by vegetation, the general layout should be planned rationally with the buildings laid out beyond the tree influence radius, and light trans­piration trees should be selected for afforestation (see Table 3).

2. A lime trench may be dug between the house site and the tree before construction to create a

-iiii

Chen and Tian 75

TABLE 6 The Calculated Foundation Soil Shrinkage Rate of Different Regions

Region

Yun Gui Index Mengzi Panxi Kunming Qujing Wenshan Ningming County Linyi County Hefei Jingmen Chengdu Nanjing Nanning

CFS(%) 81 64 62 80 52 68 53 61 50 64 64 61 57 56

~ 34 25 32 28 27 29 22 29 35 23 24 21 21 29 m 0.60 0,60 0.70 0.70 0.70 0.74 0.74 0.76 0.80 0.80 0.80 0.90 0.90 0.87

•o 1.15 1.04 1.16 1.13 1.21 0.79 0.63 0.83 0.92 0.65 0.60 0.64 0.65 0.89 Cm 0.70 0.63 0.67 0.59 0.71 0.58 0.56 0.59 0.76 0.54 0.51 0.57 0.55 0.70 Si 21.0 20.0 22.7 25.0 22.6 11. 7 4.3 13.l 8.3 6.7 5.6 4.3 6. 1 10.0

• • • • • • + + + + + + • Note: Mark"•" means damage to houses is related to tree innuence. Mark 0 +" means damage is not a result of tree influence.

KM 0.9

0.7

0.5

s, 10"/o

• No damage

x Damaged

I- ---x-- KM0.9

• •. : : L. I

~x x x x xxx

filled with lime powder

___ _.. __ _... ___ .._ __ _._ ___ .... __ ...... s, FIGURE 17 Schematic of lime trench separation method.

0 5 10 15 20 25 30

FIGURE 15 Relationship between Km and S1.

• No damage

S1 10% x Damaged I x

1.4- I x

I x x I x .. , x xx x

xx x x .. )( • : I x x I x

x e0 0.7

1.0

L- ----- - -----0.5

.. . : .. ·~ .· ... _ ..... ___ .._ __ ..... __ ...., ___ .... __ .... _s, 0 5 10 15 20 25 30

FIGURE 16 Relationship between e0 and S1.

moisture transfer barrier. The depth of the trench should be 2 m and the length should be equal to the tree influence radius. One side of the trench should be filled with lime to the ground surface; the thickness of the lime fillings should be 10 to 20

cm. The other side of the trench, after having been filled with earth, should be ranuned and sealed with cement (Figure 17). Whether using this method or cutting down the tree to build the house, the land should not be used for construction for 0.5 to 1 yr. The house cannot be built until the humidity varia­tion of the foundation soil is balanced and becomes stable.

The following trees are suitable for affores­tation:

Evergreen Trees Pine, Cypress

Deciduous Trees Arbor: Chinaberry, Dawn redwood, Gingko Bush: Chinese scholartree, Rose Shrub: Cottonrose, Crape Myrtle

.CONCLUSIONS

The following conclusions can be drawn:

1. Transpiration by water suction of tree roots causes partial shrinking deformation of foundation soil, which causes substantial differential settle­ments of engineering structures and resulting damage. This is the result of the effect of vegetation on foundation soil.

2. The radius and depth of the sphere in which the vegetation transpiration affects foundation soil is dependent on the species, age, and the growing condition of the roots of the tree. The influence radius of a grown-up tree is 1. 4 times the tree height, and its influence depth is 3. 5 to 8 m. The degree of deformation will become greater closer to the tree.

3. The effects of vegetation transpiration on dry high void ratio expansive foundation soil mainly occurs in dry and sub-dry climate areas. In the area in which Km> 0.9, e 0 < 0.7, or s1 < 10 percent, the vegetation effects are not significant.

REFERENCES

1. N,J. Rothenberg and B.L. Blace. Microclimate-­Biological Environment. John Wiley and Sons, New York, 1983.

2, Hexinfang and c. Lin. Moisture Coefficient of Expansive Soil and Its Application in Engineer­ing Practice. Fifth National Session of Expan­sive Soil, 1984.

Publication of this paper sponsored by Conunittee on Environmental Facto~s Except Frost.